[07:01] duflu: Hi :) Got 2 minutes ? [07:03] Hello MCR1, yes? [07:03] duflu: Enabled Framebuffer object causes laggy and stuttering graphics if unityshell is enabled with recommended gallium driver at least. Is there a reason to turn it on by default ? [07:04] MCR1: Yes, because with most graphics configurations, the opposite is true. I have tested 6 different graphics drivers to come to that conclusion [07:05] Or do you mean /only/ with unityshell? [07:05] yes [07:05] Hmm [07:05] It is 100% reproducable [07:05] Yes I suspect the new FBO logic in compiz is not optimally integrated with that of Nux (Unity) [07:05] There may be some inefficiency there [07:05] And the result is highly visible [07:05] I have just tested AMD/gallium [07:05] I don't have trunk Unity built to verify right now but will test it as soon as I do [07:06] Thx, because I think this would be quite important for every AMD/ATI user. [07:06] MCR1: Make sure you have this fix: https://bugs.launchpad.net/bugs/1064834 [07:06] Ubuntu bug 1064834 in nux (Ubuntu) "Compiz spends 94% of its CPU time in glTexImage2D() from nux::IOpenGLSurface::InitializeLevel() from ... nux::GraphicsEngine::CreateTextureFromBackBuffer when the dash is open" [High,Triaged] [07:07] MCR1: Please also log a bug in case I forget [07:08] duflu: I have this fix afaik, I am on trunk of Compiz/Unity and tested again yesterday... [07:08] duflu: Ok, I'll file a bug. [07:08] MCR1: You also need trunk lp:nux [07:09] duflu: I make sure I have tested and verified that before filing the report. Thx 4 your time. [07:10] hm, i can't find the unity --reset bug but i think phanimahesh has a python script that does something similar in a dirty way. [07:12] Here's the script http://etherpad.ubuntu.com/kftNoOVFYM [07:12] in case it doesn't work, http://paste.ubuntu.com/1280650/ [08:18] Mirv: sil2100: hey! :) [08:19] Mirv: sil2100: are you preparing the SRU to get it into -proposed tomorrow, as we planned? [08:40] didrocks: very much so, the problem is that getting the fixes done was a bit delayed so we'll need to push hard to get also testing done on time [08:40] (and we need SRU acks) [08:41] ok :) [08:41] SRU acks? [08:48] ie. following https://wiki.ubuntu.com/StableReleaseUpdates#Procedure [08:53] Mirv: Hi :) I want to get bug 1063249 into SRU for Precise and Quantal - The fix has no regression potential and is tested and merged. Should I just subscribe the team ubuntu-sru ? [08:53] Launchpad bug 1063249 in unity (Ubuntu) "Unity Show Desktop fade back in function not smooth" [Undecided,In progress] https://launchpad.net/bugs/1063249 [08:54] MCR1: yes please, just change the bug description to have the required items from the above wiki page, and then subscribe [08:54] Mirv: ok [09:01] Mirv: Done. :) [09:02] MCR1: could you also submit a merge request for lp:unity/6.0 ? [09:02] sil2100: Good you are here - I need you for 1 min. [09:03] sil2100: Will do, but another question: Can it be that the automerger does not like https://code.launchpad.net/~mc-return/unity/unity.merge-fix1006429-fix1006434-fix1063171/+merge/129150 ? [09:03] hey, i am curious if there was a need to put back unity --reset. me and my friends put together a python script that does it somewhat. [09:04] sil2100: You know it was the branch accidentially merged... [09:04] here's the script - http://paste.ubuntu.com/1280650/ [09:04] sil2100: I had to re-commit my one line fix and repropose to merge... [09:05] sil2100: Maybe that is why it does not land - if so I can ofc apply the fix once again to trunk and propose a new merge... [09:05] sil2100: This fix is closely connected to other showdesktop bugs, so I would really like to get this in also. [09:19] sil2100: I am not entirely sure how to best submit a merge request for lp:unity/6.0 - Should I check out this branch and then apply the fix, commit, push and request the merge or is there an easier way ? [09:24] MCR1: one moment, looking [09:24] thx [09:24] Mirv: sil2100: there is still the nux issue on armfh/armel to build, wasn't it supposed to be fixed? [09:25] didrocks: I read an e-mail from Martin today that he sent on Friday that it seems 'not fixed' even though it got marked as resolved [09:25] didrocks: so the nux guys need to check it out again [09:26] sil2100: the issue is that nux FTBFS on armfh and so no install is possible? [09:27] MCR1: about the 6.0 merges - yes, sadly, you have to branch lp:unity/6.0, apply your changes, commit, push and request - best if its also tested before submitting if it works with the 6.0 code-base [09:28] didrocks: that was the issue before, yes, so I suppose its still the same problem - as Martin posted a jenkins log on where he reproduced the nux build failure on Friday === mmrazik is now known as mmrazik|afk [09:28] sil2100: Ok :-( - although this particular fix will work in Unity 6.0 for sure as this bug is old and the showdesktop unity code has not changed for a long time... [09:28] sil2100: ok, please keep me in touch [09:29] sil2100: What about https://code.launchpad.net/~mc-return/unity/unity.merge-fix1006429-fix1006434-fix1063171/+merge/129150 ? [09:29] didrocks: but now that I see it, it's just a bunch of unused variables with -Werror=unused-parameter ... strange thing [09:29] MCR1: looking at this one right now [09:30] sil2100: oh, sry [09:30] did not want to stress [09:30] lp:unity uses the new merger, so not sure why its stalled since Friday - let's try re-approving it [09:31] sil2100: thx, I am preparing the fix for 6.0 [09:31] MCR1: thanks! Will keep you in touch about the one-liner-fix ;) [09:31] thx [09:43] sil2100: Done (6.0 merge request): Is this okay ? : https://code.launchpad.net/~mc-return/unity/unity6.0.merge-fix1063249/+merge/129632 [09:46] MCR1: thanks, looking good! [09:46] MCR1: will test and approve ASAP [09:46] sil2100: You can ofc test, but you do not need to. [09:47] * MCR1 is preparing the other showdesktop fix (requirement) for Unity 6.0 as well... [09:48] aeh - feature, not requirement [09:56] sil2100: https://code.launchpad.net/~mc-return/unity/unity6.0.merge-fix1006429-fix1006434-fix1063171/+merge/129636 [09:57] sil2100: This should also be part of the SRU, but it has 3 bug reports linked it fixes. [10:06] MCR1: first, I need to check why its not getting picked up by the merger for trunk ;) [10:08] sil2100: If it creates problems, just tell me and I will apply the fix to current trunk and propose a new merge request :) [10:16] hi, maybe someone can help me a little bit. I'm having difficulties settings shortcuts in quantal without a modifier. E.g. some notebook's special keys. They get recognized in the gnome settings but then are ignored. I'm pretty sure I have seen a similar bug report once but cannot find it anymore. Maybe someone can point me in the right direction?! === MacSlow is now known as MacSlow|lunch === mmrazik|afk is now known as mmrazik [11:17] mmrazik: hi! [11:18] sil2100: Oh - problem ? [11:20] MCR1: it seems your fix has some fatum on it, maybe some curse or something - but we're tracking it [11:22] MCR1: ah, but Martin just fixed it, it should be merging pretty soon [11:22] sil2100: Probably because it is the same branch the merger thinks he already merged it... Ah - good news as I've no work to do then :) [11:23] Thanks, mmrazik :) [11:25] sil2100: Last question for today: Do you know why latest nux is not available in any of the usual PPAs (staging, stagingfuture) and do you know if this will change soon ? [11:36] MCR1: I see staging has the latest nux in it [11:36] MCR1: revision 688 from lp:nux [11:36] MCR1: but we have problems with nux arm builds currently [11:36] (FTBFS for arm*) [11:42] sil2100: You are right - just apt chose 3.8.0+bzr+pkg368~quantal1 here, because it thinks this version is higher than 3.8.0bzr688pkg368quantal6 - probably because of the missing "+" between 3.8.0 and bzr... === _salem is now known as salem_ === dandrader is now known as dandrader|afk === dandrader|afk is now known as dandrader === MacSlow|lunch is now known as MacSlow === dandrader is now known as dandrader|afk === Guest98741 is now known as balloons === balloons is now known as Guest27823 === dandrader|afk is now known as dandrader === Guest27823 is now known as balloons_ === balloons_ is now known as balloons === yofel_ is now known as yofel [14:33] mhr3, hey [14:34] davidcalle, hi [14:34] mhr3, how are you doing? [14:42] I need some help with the python gsettings api [14:45] davidcalle, preparing myself mentally for next week :) [14:56] mhr3, :) [14:56] mhr3, is it still possible to specify a category header for the home Dash, by default now we have have the lens name and the icon of their first category. [14:56] davidcalle, what about you? [14:57] mhr3, I'm pretty good :) [14:57] davidcalle, nope, it's the lens name atm, no way to override [14:58] mhr3, meh [14:58] mhr3, ty :) [14:59] davidcalle, well, there is the property to set it, but i think unity doesn't honour it [14:59] it might in the future though [15:05] mhr3, it's for an invisible lens, the calculator/weather/'n various stuff. I think I'll split everything into lenses. By the way, Unity also doesn't care about the personal_content hint sent from Python. [15:07] are you sure? [15:08] mhr3, well, every scope using it should appear above the shopping one, right? === salem_ is now known as _salem [15:16] well, unless there are already results [15:17] actually, i'm not sure now [15:18] i think even non-personal results would show up above shopping if there's space [15:18] * mhr3 looks at the code [15:23] mhr3, not sure what you mean by "if there's space" [15:23] ie there are no other results [15:25] mhr3, if I make a search returning only photos and shopping results, I consistently have the shopping cat on top, even if it appears after the photos cat. [15:35] davidcalle, looking at the code, shopping should be moved to 3rd place everytime there are some results [15:37] all scopes need to properly emit finished for it to work though [15:43] mhr3, haha just found the issue : I had "scope.provides_personal_content=True" instead of "scope.props.provides_personal_content=True" [15:44] heh, python and its stupid props :P [15:49] davidcalle, still, even without the flag, the results should be above shopping, if only photos and shopping is visible for the search you did [15:50] mhr3, without the flag, shopping is above every Python lenses [15:50] then there's a bug somewhere :) [15:51] :) [15:51] and if you find where, i'll make sure it gets sru-ed === _salem is now known as salem_ === Guest58037 is now known as jbicha_ === dandrader is now known as dandrader|lunch [16:46] bschaefer: Hi :) If you have a minute - this one needs approval (I guess it does not need any testing, just a look): https://code.launchpad.net/~mc-return/unity/unity.merge-fix-xml.in-headers/+merge/129713 [16:46] MCR1, ill take a look in a bit :) (Thanks!) [16:47] bschaefer: Thx a lot also :) [16:52] MCR1, since I don't know a lot about XML, is just the header that XML files need? [16:58] bschaefer: Yes, this header is partially already in use for .xml.in files, not only in lp:unity, but also in lp:compiz. Please take a look at the 2 screenshots I attached to the linked bug report as well :) [16:59] MCR1, cool, will do [16:59] Oh, Trevinho was fast :) Thx 4 approval. :) [16:59] bschaefer: Thx also :) [16:59] MCR1: yw ;) [16:59] MCR1, looks like marco beat me to it :) [16:59] Trevinho, thanks [17:15] Trevinho: Thx for merging my triple bug fix to Unity 6.0, but I am not sure what happens if someone already has fadetodesktop or showdesktop activated -> might be that those have to be deactivated manually then - that is why I have not proposed the fix for 6.0 yet... [17:17] Trevinho: I mean if those computers explode, you have been warned ;) === dandrader|lunch is now known as dandrader === Guest72485 is now known as dpb___ === jbicha_ is now known as jbicha === salem_ is now known as _salem === _salem is now known as salem_ === dandrader is now known as dandrader|afk [20:06] Another attempt at fixing unity --reset, using Python Gobject bindings. http://paste.ubuntu.com/1281827/ [20:06] The revision history is saved on bitbucket at https://bitbucket.org/jpmahesh/unity-reset [20:08] phanimahesh, unity --reset is no more, it is just unity (which does the same thing now) [20:10] I'm not on quantal yet, but there has been discussion that unity can not be reset to defaults as of now. [20:11] And this is supposed to fix that. === dandrader|afk is now known as dandrader [20:11] o hmm, cannot be reset to defaults? As in the default compiz profile? [20:12] yup. stash all customisations. [20:12] which actually sounds like a good idea [20:13] awesome, but didrocks would be the one to talk to, about getting it in :) [20:14] But I'ven't tested it yet. If anyone is on Quantal, can you change launcher icons size, autohide, and such settings at random and see if running this resets everything? [20:15] hmm let me see if it'll work for me :) [20:15] I pinged Jorge catro. He'll talk to didrocks, [20:15] cool [20:15] The first version was dirty, and didrocks commented that it'd be better to rewrite it to use python bindings, and i just did that. [20:16] in the paste bin you have up there? [20:17] the pastebin one is the newest version. [20:17] old ones are in the git repo, as separate files. [20:18] cool, Ill give it test [20:23] phanimahesh, when I set the launcher to autohide, it doesn't reset to default [20:23] can you pull up a python shell? [20:24] yeah [20:25] oops, fixed it. I've not imported the library. wonder hw it ran in the first place. [20:25] hm it must have just skiped to the unity --replace === salem_ is now known as _salem [20:26] can you pull up the updated file reset-gio.py from the repo. [20:27] phanimahesh, also if I just run that script...it doesn't seem to call anything but the unity --replace === _salem is now known as salem_ [20:27] I've made a couple of really really bad errors. [20:27] yup will do [20:27] yup, just noticed that I missed the function calls and import. fixed them. [20:27] awesome :) [20:28] so... it works? [20:28] hmm the launcher still wants to autohid [20:29] im not sure where it would change the autohide back from 1 to 0, it seems you just use a list of plugins [20:29] the autohide should be somewhere under... [20:30] it should be under the unityshell plugin [20:30] com.canonical.Unity.Launcher [20:30] nope. this one has it. [20:30] and I did reset it in my script. [20:30] lets try it manually. can you open up a python shell? [20:30] hmm [20:31] yup [20:31] from gi.repository import Gio [20:32] * bschaefer installing ipython really quick [20:32] cause im lazy [20:32] alright [20:33] python -c 'from gi.repository import Gio;print Gio.Settings(schema="com.canonical.Unity.Launcher").list_keys()' [20:33] in terminal. [20:33] what does the output read? [20:34] ['favorite-migration', 'favorites'] [20:34] * bschaefer wonders where that moved to [20:36] this, is something wierd. :P [20:36] can you look up the autohide setting in dconf-editor and find its schema? [20:36] yeeah, Ill see if I can find where it is [20:37] yeah [20:40] got it? [20:41] hmm I dont see it under com.canoncial.Unity [20:45] the only place I can see it under the gconf-editor is under my profile/plugins [20:45] okay. when you see it, select it and look at the bottom portion. it should tell you the schema. [20:46] it says No Schema :( [20:46] :( [20:46] whats the key? [20:46] autohide ? [20:46] launcher-hide-mode [20:46] hmm odd [20:46] smspillaz, would know [20:47] but he wont be around for a bit [20:48] okay. gimme a sec. [20:49] this is were it is declared: http://bazaar.launchpad.net/~unity-team/unity/trunk/view/head:/plugins/unityshell/unityshell.xml.in#L50 === salem_ is now known as _salem === _salem is now known as salem_ [21:00] Thanks. [21:01] python -c 'from gi.repository import Gio;print Gio.Settings(schema="org.compiz.unityshell").list_keys()' [21:10] (process:1198): GLib-GIO-ERROR **: attempting to create schema 'org.compiz.unityshell' without a path [21:10] Trace/breakpoint trap (core dumped) [21:10] wait [21:10] * bschaefer tries something [21:11] nevermind I get a seg fault [21:11] when I try that...hmm [21:16] bschaefer: I've cleaned up the mess in unity/plugins/unityshell.xml.in - 1 more tab, better structuring, better indentation, easier readability, improved explanations/tooltips - I would be pleased if you could try: https://code.launchpad.net/~mc-return/unity/unity.merge-unityshell-compiz-ccsm-plug-in-refactoring/+merge/129724 [21:16] MCR1, awesome [21:16] * bschaefer looks [21:16] bschaefer: The diff looks unreadable now ofc. :-[ [21:17] bschaefer: I guess you have to try it... [21:17] MCR1, was the refactoring it just adding spaces and making it more readable? [21:17] bschaefer: This is something I wanted to do for a long time, because it was a big mess... [21:17] * bschaefer pulls the changes [21:18] bschaefer: Nope, it is quite a bit more - but no options have changed, just what I wrote above. [21:19] bschaefer: Essentially this fixes nothing, but should make the unityshell Compiz plug-in much more structured... [21:19] MCR1, awesome, I think the design team would have to look at the string changes through... [21:20] like the tooltips [21:20] bschaefer: Those just replace "weak" English with a bit better (also weak) English, but you could proofread them as you are a native speaker... [21:21] MCR1, yeeah about being a native speaker != strong english :) [21:21] but I can proof read it :) [21:21] I just fixed obvious grammatical mistakes and added a "." to the end of the long descriptions/tooltips... [21:22] To unify the look of them - simply does not look professional to have one tooltip end with "." and the other without... [21:23] cool, I don't think it'll need the design team [21:24] Important is that keys are now on top of each tab and options are sorted - New tans are "General", "Dash", "Launcher", "Switcher", so it is now easy to find and change things... [21:24] *tabs [21:25] cool [21:25] yeah, I usually just grep to find what im looking for [21:26] and it is more easy for a human to read the unityshell.xml.in file now, because indentation was completely off before... [21:28] yay, readability is very important [21:28] After all humans are no machines ;) [21:29] yet! [21:29] hehe [21:29] Unity brain extension already planned ? [21:29] I was hoping more for cyborg stuff [21:29] or aiming more at [21:30] Play UFO:AI then ;) [21:30] * bschaefer has no time [21:30] oh wait - they do not have implemented it yet ;) [21:31] haha [21:34] MCR1, 46 +